Staphylococcus aureus is the leading cause of cardiovascular device infections (CDI)(61,203,205). Adhesion to surfaces is the determining first step in staphylococcal infections(82). Receptors that bind human fibronectin (Fn) and fibrinogen (Fg) are critical because these are two host proteins that coat implanted devices in vivo (83,104,199-201). Cell wall anchored Fn-binding (FnBPA, FnBPB) and Fg-binding (ClfA, ClfB) proteins on S. aureus are associated with cardiac device infections in vivo(6,51,150,157,159,188). Our hypothesis is that single amino acid polymorphisms in FnBPs and Clfs alter the binding properties of these adhesins and are directly linked to the etiology of S. aureus infections of cardiovascular implants in humans. We recently completed a R21 grant that demonstrated proof of principle as confirmed by peer-reviewed papers in Proc Natl Acad Sci USA (117) and J Biol Chem (27). For the R21, we examined 46 bloodstream isolates of S. aureus from patients with infected (CDI) or uninfected (CDU) cardiac prostheses. We discovered an association (p <0.01) between CDI and three distinct, non-synonymous single-nucleotide polymorphisms (SNPs) in fnbA corresponding to E652D, H782Q and K786N in FnBPA. Quantitative atomic force microscopy (AFM) measurements on living cells revealed a stronger binding force (p <0.0001) for CDI (n = 26) vs. CDU isolates (n = 20). Complementary molecular dynamics (MD) simulations showed extra hydrogen bonds for each polymorphic amino acid. This R01 describes an innovative collaboration from the bedside to the bench that tests the molecular basis of altered binding through a combination of clinical studies, high-throughput genome sequencing, and state-of-the-art instrumental and computational analyses. In Aim 1, we will determine variations in known microbial surface components recognizing adhesive matrix molecules (MSCRAMMs; e.g., fnbA, fnbB, clfA, clfB) by whole genome sequencing of >150 S. aureus clinical isolates obtained from the bloodstream, nares, and explanted device of patients with cardiac implants, and correlate the data with clinical outcome of patients. In Aim 2a, we will characterize the significance of MSCRAMM variants at the cellular level by using AFM to directly probe the strength of ligand-receptor bonds on living Lactococcus lactis, a surrogate Gram-positive, that express recombinant FnBPs or Clfs with binding motifs that mimic statistically significant SNPs identified in Aim 1. In Aim 2b, Surface plasmon resonance (SPR) and isothermal titration calorimetry (ITC) will be used on a select number of polymorphic proteins from L. lactis that show significant distinction in binding at the cell level. Synthetic peptides mimicking key SNPs will also be tested with AFM, SPR, ITC, and MD simulations. In Aim 3, we will confirm the molecular basis of altered binding by using AFM to directly analyze polymorphic-MSCRAMMs as they are expressed in their respective clinical isolates or S. aureus mutants lacking fnb & clf. This work will provide a fundamental understanding of the initial step in S. aureus CDI, and identify novel ways to prevent bacterial infections of cardiac implants.